GALACTOSEMIA: THE SOUTH AFRICAN STUDY
I
-PHIYANI JUSTICE LEBEA
MSc (Biochemistry, PU for CHE)Submitted in fultilment of the academic requirements for the degree
PHILOSOPHIAE DOCTOR
in the
School of Biochemistry Faculty of Natural Science
North-West University Potchefstroom Campus
Potchefstroom South Africa
Promoter: Prof P.J. Pretorius Co-promoter: Dr M. Weinberg
For my loving parents
God had a wonderfir1 advantage of usorking alone KO$ Anan
ABSTRACT / SUMMARY
Classic galactosemia and congenital hypothyroidism are both inborn errors of metabolism. The biochemical and molecular detection of both disorders is already defined in scientific literature. Their management is also well documented, yet their detection especially in poorer communities remains uncoordinated moreover in developing countries. These disorders can both be detected early in childhood if neonatal screening is instituted. When diagnosed early in childhood, congenital hypothyroidism can be successfully treated. In the case of classic galactosemia, preliminary clinical outcome is satisfactory although long-term prognosis is disappointing.
In South Africa, neonatal screening has not been instituted. The reasons are related to the economics of the country as well as the lack of sufficient data about the prevalence and incidence of the diseases.
In this study, a pilot newborn screening program for classic galactosemia and congenital hypothyroidism that take into consideration the socio-economic status of the surveyed populations has been established. The total protocol includes the sampling, storage of sampled specimens, biochemical and molecular diagnostic techniques. Using this protocol, an incidence ratio of one in a thousand congenital hypothyroidism cases was established in the Nkangala region of the Mpumalanga province of South Africa. Appropriate management therapy was also instituted in affected individuals.
Furthermore, to initiate a lead into the investigation of mechanistic imperatives that result in poor long-term prognosis of classic galactosemia, a novel approach towards application of RNA interference in disease mechanism study was introduced. In this approach, the design and development of a mammalian cell model with GALT gene knockdown using RNAi was utilized to create a cellular state reminiscent of a classic galactosemia rather than to elucidate the gene function as is conventionally applied. The model was successfully completed and was compatible with the enzymatic activity prerequisites when compared to the control sets. However, the mammalian cellular model still needs to be rigorously tested to confirm its application towards studying long-term biochemical outcomes and identification of galactosemia secondary biomarkers.
OPSOMMING
Klassieke galaktosemie en kongenitale hipotiroiedisme is beide aangebore metaboliese siketes. Die biochemiese en molekulere waarneming van beide siektetoestande sowel as hul behandeling, is reeds goed beskryf in die wetenskaplike literatuur. Tog is die aantoning van hierdie siektetoestande in armer gemeenskappe ongekoordineerd, meerso in ontwikkelende lande. Indien kongenitale hipotiroiedisme vroeg in 'n kind se lewew gediagnoseer word, kan dit suksesvol behandel word en in die geval van klassieke galaktosemie is die aanvanklike kliniese uitkoms bevredigend maar die lang temyn prognose is egter nie na wense nie.
Neonatale sifting is tans nog nie in Suid Stiika ingestel nie, heelwaarskynlik om ekonomiese redes maar ook as gevolg 'n gebrek aan data ten opsigte van die voorkoms van hierdie siektetoestande.
In hierdie studie is 'n lootsprogram vir die neonatale sifting vir klassieke galaktosemie en kongenitale hipotiroyedisme ontwikkel waarin veral die sosio-ekonomiese aspekte van die bevolkings wat betrokke was, in aggeneem is. Die volledige protokol sluit die volgende in, nl. die neem en berging van bloedmonsters en die biochemiese en diagnostiese tegnieke. Toepassing van hierdie protokol het 'n insidensie van een in 'n duisend vir kongenitale hipotiroiedisme in die Nkakala-distrik in die Mpumalangaprovinsie in Suid Afrika aangetoon. Geskikte behandeling is ook vir die betrokke individue ingestel.
Ten einde 'n ondersoek aan te voor na die meganismes wat verantwoordelik is vir die swak lang tennyn prognose van klassieke galaktosemie, is 'n unieke benadering gevolg ten opsigte van die toepassing van RNA-tussenkoms (RNAi) in die betudering van die meganisme van hierdie siektetoestand. In hierdie benadering is die ontwerp en ontwikkeling van 'n soogdier selkultuumodel ontwikkel waarin die GALT-geen onderdruk is om toestande te skep soortgelyk aan klassieke galaktosemie, eerder as om die konvensionele weg van die opklaring van geenfunksie te volg. Die model is suksesvol toegepas en was goed vergelykbaar met die vereistes wat gestel is ten opsigte van ensiemaktiwiteit in vergelyking met kontrolestelle. Hierdie model moet egter nog verder uitgebou en aan streng toetsing ondenverp word ten einde die toepasbaarheid daarvan op die bestudering van die lang termyn biochemiese uitkomste van klassieke galaktosemie en ook vir die identifikasie van sekondere biomerkers vir hierdie siektetoestand.
TABLE OF CONTENTS
...
SUMMARY .i . . OPSOMMING.....
.II...
LIST OF ABBREVIATIONS.. vi LIST OF FIGURES...
x LIST OF TABLES...
xi LIST OF APPENDICES...
xi ACKNOWLEDGEMENTS.....
..xii CHAPTER ONE GENERAL INTRODUCTION.....
.
.
...
...
...
1.1. Scope and structure of the study.....
1.2. Study area and population.....
CHAPTER TWO
CLASSIC GALACTOSEMIA AND CONGENITAL HYPOTHYROIDISM
IN SOUTH AFRICA
...
2.1. Introduction
...
2.2. Classic galactosemia.....
...
2.2.1. Biochemical and molecular diagnosis of classic galactosemia..2.2.2. Treatment and management of classic galactosemia.
...
2.3. Congenital hypothyroidism.....
...
2.3.1. Biochemical diagnosis of congenital hypothyroidism..2.3.2. Treatment of congenital hypothyroidism..
...
2.4. Perspective over newborn screening in the South AfricanSOCIO-econom~c environment..
...
2.5. Logistical challenges and cost efficiency of newborn screening for congenitalhypothyroidism and classic galactosemia in a rural South African settmg..
...
2.6. A novel approach for improved qualitative SNP genotyping designed for low starting template DNA concentration in real-time
PCR
...
32 2.7. Conclusion...
46CHAPTER THREE
POSSIBLE REASON FOR POOR LONG-TERM PROGNOSIS IN CLASSIC
GALACTOSEMIA PATIENTS
...
3.1. Introduction...
3.2. The molecular relationship between deficient UDP-galactose uridyltransferase (GALT) and ceramide galactosyltransferase (CGT) enzyme function: A possible cause for poor long-term prognosis in classic
...
galactosemia
CHAPTER FOUR
RNA INTERFERENCE. MAMMALIAN CELLS AND GALACTOSEMIA
CELLULAR MODEL DEVELOPMENT
...
4.1. Introduction...
4.2. Bacterial model for classic galactosemia...
...
4.3. Yeast model...
4.4. Mouse model4.4.1. High galactose concentration loading model
...
4.4.2. GALT knock-out mouse model...
4.5. Human galactosemia cellular model...
4.6. The human cell RNAi induced galactosemia model...
4.7. Materials and methods...
4.7.1. Design of shRNAs and target identification...
....
4.7.3. Mammalian cell lines, culture conditions and antibiotic resistance
...
734.7.4. Transfection conditions for shRNA expressing plasmid
...
744.7.5. Phenotypic assays after transfection
...
754.8. Results and discussion
...
774.9. Conclusion
...
83CHAPTER FIVE GENERAL DISCUSSION. CONCLUSION AND RECOMMENDATIONS 85 5.1. South AGican legislation and congenital birth defects
...
855.2. Screening for galactosemia and hypothyroidism in newborn infants
...
86...
5.3. The RNAi system and galactosemia 87 5.4. References...
89ABREVIATIONS Ago AIDS BDSS bp BSA
O
c
Cat no. CDC cDNA CGT CH CNSco2
CRC ddn2O -dF/dT DOH DOL DMSO DNA dsRBD dsRNA E coli EC ECM EDTA EIA Et al. ArgonuateAcquired Immunodeficiency Syndrome
Birth Defects Surveillance System base pairs
Bovine serum albumin
degrees Celsius Catalogue number
Centre for Disease Control and Prevention, Atlanta, Georgia.
complementary DNA, obtained by reverse transcription of the mRNA ceramide galactosyltransferase
Cogenital Hypothyroidism Central Nervous System Carbon dioxide
The United Nations Convention on the Rights of the Child
double distilled water
negative change in fluorescence as a factor of change in temperature South African National Department of Health
South African National Department of Labour Dimethylsulfoxide
Deoxyribonucleic acid
double-stranded ribonucleic acid binding domains double-stranded RNA
Escherichia coli
Enzyme Commission Number: denotes a specific number that an enzyme is given based on function and structure
Extracellular matrix
Ethylenediamine tetra-acetic acid: Ci&N208 Electroimmunoassay
EtBr Ethidium bromide: Ci2H2~BrNa FCS FSH fwd gal gal-l-p GalC GALE GALK GALT GFP L Lysis buffer
Fetal Calf Semm
Follicle Stimulating Hormone Forward primer Galactose Galactose-l-phosphate Galactosylceremide Uridinediphosphategalactose Cepimerase Galactokinase
Galactose-l-phosphate uridyl transferase Green fluorescent protein
Water
Cells which were originally obtained from the cervical cancer cell line of Henrieta Lacks
Human Immunovirus hours
Human uridylgalactose phophorylase 2
Intelligence Quotient
kilograms
litre
20mM Hepes-KOH, pH 7.5 I ImM DTT with BSA at 0.3mgiml
microlitres micromolar
micromole per hour per gram haemoglobin microgram
Molar: moles per litre milligrams
Magnesium ion Magnesium chloride milligram per decilitre minutes
mol m U L mRNA MW NAD+ NADH NADP' NADPH NGO NaHCO, ng nm nM 0 OH OMIM % PAZ PBS PH PHC PKR PNS microRNA
micro ribonucleoprotein particle millilitres
millimetres millimolar moles
milliunits per litre messenger RNA Molecular weight
Nicotinamide Dinucleotide (oxidised form) Nicotinamide Dinucleotide (reduced form)
Nicotinamide Dinucleotide Phosphate (oxidisedform) Nicotinamide Dinucleotide Phosphate (reduced form) Non-Governmental Organisation Sodium carbonate nanogram nanometres nanomolar Oxygen Hydroxyl
Online Mendelian Inheritance in Man
percentage
Piwi, Argonuate and Zwille protein domains Phosphate Buffered Saline
a measure of acidity numerically equal to the negative logarithm of H' concentration expressed in molarity
Primary Health Care
double-stranded RNA-dependent protein kinase Peripheral Nervous System
Quantitative Reverse Transcriptase Polymerase Chain Reaction
Repeat-associated short interfering RNAs RNA-dependent RNA polymerase
...
Ref rev RFW RISC RNA RNAi RNP RT-PCR T4 TBE TE buffer TNR TRAIL TSH U UDPgal UDPglu UTR Reference Reverse primer
Relative Fluorescence Unit RNA-Induced Silencing Complex Ribonucleic acid
Ribonucleic acid interference Ribonucleoprotein particles
Reverse Transcription Polymerase Chain Reaction
Standard deviation Short interfering RNA Short hairpin RNA
Thyroxine
Tris Borate EDTA buffer: 89.15 nM (pH 8.0), 88.95 mM boric acid. 2.498 mM Na2EDTA
Tris EDTA buffer: 10 mM Tris HCI (pH 8.0). 1 mM Na,EDTA Tenascin-R
Tumor Necrosis Factor Apoptosis Inducing Ligand Thyroid Stimulating Hormone
Units of enzyme activity Uridinediphosphogalactose Uridinediphosphoglucose Untranslated region
volume per total volume
LIST OF FIGURES
Figure 1.1: The map of Southern f i c a showing the location of Nkangala region in Mpumalanga
...
province and North-West University laboratory .5
Figure 2.1: The schematic representation of the Beutler method of GALT activity assay
...
1 I Figure 2.1 : Graph of total galactose versus sample frequency and cumulative frequency.....
.34 Figure 2.2. Graph of TSH concentration versus sample frequency and cumulative frequency.. .34 Figure 2.3: Panel A; The SYBR Green I1 graph showing successive amplification cycles with theresultant reduction in fluorescence with fluorescence measured at the elongation step of the PCR cycle. Panel B; The SYBR Green I1 graph showing the presence of the amplification fragment with its melting peak at 87.5 OC in an inverted fashion.
...
43 Figure 2.4: The SYBR Green I1 graph with fluorescence measured at denaturation step, depictingthe presence of the desired amplified product at 87.5 'C and primer dimers of the
0
blank sample at 80 C
...
44 Figure 2.5: Panel A; The SYBR Green I1 and fluorophore specific hybridisation probes graphshowing the presence (green arrow) of the S135L fragment wild type peak at 55.5 OC and the absence of the mutant at 62.5 OC. Panel B; The SYBR Green I1 and
fluorophore specific hybridisation probes graph showing a heterozygote profile.
...
.,.45 Figure 2.6: Panel A, The SYBR Green I1 graph showing amplification below threshold detectionlimit.
...
..46 Figure 4.1: Total cells transfected with plasmid expressing shRNAs.....
.78 Figure 4.2: A graph showing the GALT gene knockdown efficiency.....
.79 Figure 4.3: A graph showing the GALT enzyme activity after transfection.....
. A 0 Figure 4.4: HPLC chromatogram showing U6+1 plasmid evaluation of UDP-gal and ATP...
.8 1 Figure 4.5: shl chromatogram evaluation of UDP-galactose and ATP...
82 Figure 4.6: sh3 chromatogram evaluation of UDP-galactose and ATP...
82LIST OF TABLES
Table 2.1. Criteria applied to newborn screening for congenital disorders
...
9Table 2.2: Comparison of the mechanism of improved sensitivity between SYBR' Green I and I1 real-time qualitative SNP genotyping for S135L GALT gene mutation
...
38Table 2.3: Characteristics and conceneations of PCR chemical components for the genotyping
...
the GALT S135L and Q188R mutations.
Adapted from Dobrowolski e t a [ . . 2003 47 LIST OF APPENDICES Appendix A: The newborn screening results of total galactose and TSH assays...
disc Appendix B: Dual luciferase GALT knockdown results...
101Appendix C: GALT enzyme activity assay results of the transfected Hela cells
...
102Appendix D: Positions of knockdown targeted areas of the GALT mRNA
...
103Appendix E: : shRNAs names and nucleotide sequences used for transfections
...
104Appendix F: Free websites for rational siRNA design
...
104Appendix G: The Galt target sequence and the vector map of the psicheck-2 vector
...
105Appendix H: Characteristics and concentrations of PCR chemical components for the genptyping of GALT S135L and Q188R mutations
...
106ACKNOWLEDGEMENTS
I wish to express my sincere appreciation to the following persons and institutions for their valuable contributions towards the completion of this study:
My appreciation goes to all the mothers that participated in this study for kindly allowing the collection of blood specimens from their newborn babies. Sister Sannie Mahlangu, retired nursing sister, Middelburg; for her relentless dedication towards screening and collection of blood specimens. Sister Ronelle Toute whose admirable organisational, administration and person skills made it possible for this project to take off. Mr Masilela, CEO of Middelburg hospital, for his understanding and reliable insight and permitting us to continue sampling in his hospital. Sisters and medical practitioners at the Maternity, Pediatric and High care wards of the Middelburg hospital; their continued assistance and eager attitude to do better for their patients is very much appreciated.
Mr D.P. Knoll and his Biochemistry department team of the North-West University for technical help with primary screening of galactosemia and hypothyrodism. Prof Tiaan Brink, Sharlene Niewoudt and Lee for the kind gift of the mammalian cells and advice on culture conditions. Dr C.C. Bezuidenhout for allowing me to use his realtime PCR machine for genotyping experiments and assisting with technical assistance. Dr Christinah Hajinicolau of the Chris Hani Baragwanath, Paediatric ward for supplying positive specimens of galactosemic patients. Dr Jeanette Kriel of the Universitas Hospital in Bloemfontein for supply of suspected galactosemic patient specimens. Dr Jennifer Cartwright of the Johannesburg General Hospital for her encouragement and words of wisdom and helping with the clinical treatment of hypothyrodism patients. I would like to thank Mr Ian Sinclair of the NHLS for technical assistance with GALT and GALK enzyme activity assays.
Dr Marco Weinberg for his enthusiastic and encouraging approach towards RNA interference work. I am deeply grateful to have met him at the time that I did.
Prof Piet Pretorius; he was the mainstay of my academic evolution. I would like to thank him for his brilliant mentoring prowess and kind attitude with which he treated me. For shaping and sharpening my future and standing firm in my endeavour to become a better prepared Scientist in the field of Molecular Biology. I am grateful for the selfless help he offered under difficult and sometimes unbearable circumstances. Thank you for you and your family's kind support throughout the duration of my post graduate academic life. I hold you in high regard for showing me (consciously or not) how to know and understand that people can be kind only if they choose to.
Sangita Jivan, my parents, my two sisters and my nephew. I consider all of you the backbone of my family. Thank you for the support and encouragement.
I thank God almighty for putting all the people that I have met and got to know though this work and for protecting me against harm.
CHAPTER 1
GENERAL INTRODUCTION
1.1. Scope and structure of the study
Many inborn errors of metabolism can be identified and diagnosed clinically. However, some of these disorders are difficult to diagnose early enough without the biochemical confirmation. The delay may sometimes result in irreversible damage to the infant that could otherwise have been prevented with early diagnosis. This group of inborn errors of metabolism include galactosemia and hypothyroidism (Gmters et al., 2002). Early detection of each of the two disorders is critical to ameliorate the progression of the disease state. One of the most effective ways to detect these two disorders is by way of newborn metabolic screening (Beardsall and Ogilvy-Stuart, 2004; Grosse, 2005). Since these disorders are primarily genetic, the ethnicity as well as racial origin of the patient does play a role in terms of identifying the possible mutation(s) causative of the disorder. Population dynamics and socioeconomic status also play a role when considering the health options available to the parents or guardians of a newborn with such a metabolic disorder. Poorer communities tend to have limited options with respect to healthcare choices hence this study targeted newborns in a public referral institution that would cater largely for poor and middle class communities.
This study is divided into two major sections. The first part of the study (chapter two) deals with newbom screening and its implementation in a South African community. This also includes the development of a more sensitive assay for single nucleotide polymorphism detection of known GALT gene mutations.
Previous newborn screening programmes for targeting hypothyroidism and galactosemia undertaken at the Northwest University (Potchefstroom campus) failed to produce appropriate results even though a large number (over 12 000) of blood samples were collected from newborn babies. This failure was primarily due to lack of social coordination on the programme rather than laboratory set-up and biochemical protocols (Malan and Reinecke, 2006). One of the major aims of this study was thus to establish a functional and effective way of newbom screening in a typical South African setting that accommodates rural and urban communities with the rural communities in the majority. This first part was conducted as a small scale screening programme for galactosemia and hypothyroidism at the Middelburg Hospital's maternity, high care and paediahic wards. It is worth noting that public health is about saving lives, preventing disability, and improving health-related quality of life, not necessarily about financial savings (Grosse,
2005). Yet, economics plays an important role because of the need to use scarce
resources wisely. Consequently, a sub-objective to this study was to use this study as a pilot to evaluate the cost-saving as well as cost-effective intervention of screening for both congenital hypothyroidism and classic galactosemia in a typical South African population. In this context, cost savings is defined as lowest cost possible to achieve a given outcome which can be calculated relative to other interventions or to a fixed cut- off.
Meanwhile, cost-effectiveness is defined as seeking to maximise health improvements with limited resources; that is, getting good value from investments that save lives or improve health and development and not necessarily about saving money. The second part of the study was motivated by the lack of an appropriate model for classic galactosemia. The reasons for the requirement of an applicable human cellular model to study classic galactosemia are reviewed in chapter three. In chapter four, the actual development of a mammalian cell culture model that shows biochemical phenotype reminiscent of a classic galactosemia tissue is described.
This model has been developed using RNA interference to knock down the galactose-1
-
phosphate uridyltransferase (GALT) gene to as low as 91% for the messenger RNA expression and 88% at the protein level. The metabolite concentration measurements especially UDP-galactose were significantly lowered in GALT gene knockdown cells as is the case in galactosemic tissues (Lai et al., 2003). Hence the second aim of this study was to develop a mammalian cellular model that can mimic classic galactosemia tissues in terms of biochemical phenotype and complement the existing cellular models in terms of applicability. This model will be usefd to establish the link between the biochemical phenotype and the genetic defects observed in mammalian galactosemic cells and thereby explore the relationship between the two disciplines in the hture. This is in the hope that establishing the exact imbalances in the biochemical entities of a galactosemic cell may shed some light regarding the differences in the long-term prognosis of classic galactosemia patients.1.2. Study area and population
The study was conducted primarily in the Middelburg hospital which is the premier referral centre for the Nkangala region of the Mpumalanga province in South Africa. This region is inhabited by four racial groups found in South Africa; namely, White (17%), Africans (68%), Indians (9%) and Coloureds (6%). The population racial classification in this study is described according to the Statistics South Africa format of 2002. The dominant ethnic group in the Nkangala region is the Ndebele speaking people which are of an African descent. There is also a considerable number of Sepedi, Xitsonga and Siswati speaking people found in this region that are also of African descent. The samples used in this study were all collected at the Middelburg hospital since it would be representative of the Nkangala region relative to other small health centres. The percentage of patients visiting this hospital is over 90% Africans. The reason for the apparent disparity between the population demographics and the specific races visiting the centre has to do with the presence of private health clinics that would primarily cater for economically amuent and to a lesser extent the middle class communities in the area. The total population of Nkangala region is in excess of 161 000 while the birth rate in this region is 22 newborns each year per population of 1000 that is over five years of age (Stats SA, 2002). This statistic is only an indication of the ratio of population to indicate birth rate rather than population of child bearing age as per Stats SA categorisation.
Figure 1.1: The map of Southern Africa showing the location of Nkangala region in Mpumalanga province and North-West University laboratory in shades of white.
The majority of births in this region take place in conventional health institutions whether private or public. There are still considerable numbers of people that deliver their babies at home, especially those in the farm and rural areas. The farm and rural settings are inhabited by some of the poorest communities in South Africa.
The Nkangala region's geographic location is clearly outlined in figure 1.1 in relation to the laboratory in Potchefstroom (North-West Province) and the whole of South Africa. The Middelburg Hospital caters for over 40% of newborn in the region. During the study's six months sampling period, 28.9% of the newborns in the region were screened for the presence of galactosemia and hypothyroidism. The shortfall in sampling were a result of manpower shortages especially on weekends.
CHAPTER 2
CLASSIC GALACTOSEMIA AND CONGENITAL HYPOTHYROIDISM IN
SOUTH AFRICA.
2.1. Introduction
The United Nations convention on the rights of the child (CRC), of which South Africa is a signatory, compels member states to promote information exchange in functional treatment of disabled children (CRC, 1997). This includes children with genetic disorders such as galactosemia and hypothyroidism. This undertaking is aimed at enabling state parties to improve their capabilities and skills as well as to widen their experience in the relevant health area. The CRC places particular emphasis on addressing the needs of developing countries with respect to the identification, treatment and medical care of children with genetic disorders.
The South African constitution, section 27 part 2, seems to complement the CRC recommendations by including health and welfare as a basic right (DOH, 2001). According to this section of the constitution, individuals with genetic disorders, galactosemia and hypothyroidism included, should benefit from reasonable legislative and other measures that the state may introduce to achieve progressive realisation of the health and welfare right. Furthermore, the December 1997 white paper on Integrated National Disability Strategy emphasises the prevention, health care and the rehabilitation of people with disabilities (DOL, 1997). This includes prevention, health care a d o r rehabilitation that may be a consequence of a specific genetic disorder such as galactosemia or hypothyroidism.
There has been a concerted effort from the South African National Department of Health to integrate and incorporate the management and prevention of genetic disorders and birth defects into primary health care (DOH, 1999; DOH, 2001; DOH, 2004). In 2001 a booklet called policy guidelines for management and prevention of genetic disorders birth defects and disabilities, followed by another publication in 2004 delineating the available diagnostic tests in South Africa. Although the National Department of Health recognises the importance of prevention of genetic disorders, there is not as yet a large-scale screening programme that could facilitate early detection of preventable andlor manageable disorders such as classic galactosemia and congenital hypothyroidism. This study seeks to explore the applicability of such a programme cognisant of the limited resources in our country and the heavy burden of disease exacerbated by the HIVIAIDS pandemic.
The success of this part of our study would enhance efforts to improve the health and well-being of the South African populations as well as fast track the practical integration of genetic services into primary health care using newborn screening techniques. Consequently, this chapter of our study seeks to address and redress equity, efficacy, effectiveness and community participation with regards to identification and treatment of the two genetic disorders. This is in line with the described legislative requirements of the country as well as recommendations by the CRC, taking into account the need to initiate programmes that are applicable and sensitive to the economic conditions and social environment of our country's populations. The study addresses the issues pertaining to protocol development for screening of newborn infants and effective follow-up of classic galactosernia and congenital hypothyroidism (CH) patients.
The choice of the two inborn errors of metabolism to be investigated was based on the criteria defined by Van Vliet and Czernichow in 2004 as depicted in table 2.1 below. The table has been modified to include the statistics of classic galactosemia.
Table 2.1: Criteria applied to newborn screening for congenital disorders
CH Galactosemia
The disease must be known and well described
+++
+ t tThe disease must be freauent (>I : 15 000 newborns) 1 :3500 1.14 OOOa The disease must be serious and treatable ++t ++b
Patients should be identifiable before diagnosis is
suspected clinically 99% 100%
Confumatory diagnostic method exist . 100% 100%
'Data h s e d on Ojwang e r n / . . 1999 and Henderson etol., 2002 'Pmr long termprogoosis still unresolved in classic galactosemk
2.2. Classic galactosemia
Galactosemia is an autosomally recessive inherited disorder of galactose metabolism, which occurs as a consequence of a deficiency of one of the three principal enzymes involved in the metabolism of galactose, through its conversion to glucose (Kozak e t a / . , 1999; Henderson et nl., 2002). The enzymes in galactose catabolism are galactokinase (GALK, EC 2.7.1.6), galactose-1-phosphate uridyltransferase (GALT; EC 2.7.1.6) and uridine-diphosphate galactose-4 epimerase (GALE; EC 5.1.3.2). In humans, the deficiency of galactose-I-phosphate uridyltransferase produces the disorder called classic galactosemia, OMIM# 230400. In the newborn period, exposure of galactosemic infants to galactose produces hepatotoxicity, E.coli sepsis, and death in untreated patients. Survivors are known to have long term complications that include ataxia, verbal dyspraxia, and premature ovarian failure (Lai et nl., 2003).
Classic galactosemia is the most common form of galactosemia and over 170 mutations have already been discovered and described (Novelli and Reichardt, 2000).
2.2.1. Biochemical and molecular diagnosis of classic galactosemia
The diagnosis of classic galactosemia is primarily based on the combination of physical examination and laboratory results. The biochemical results would typically reveal a large number of reducing substances in urine with normal levels or undetectable glucose. Red blood cell metabolite evaluation should indicate high levels of galactose-1-phosphate coupled with reduced or non-existent GALT enzyme levels (Elsas et al., 1994; Chung, 1997).
Follow-up and monitoring of galactosemia patients is usually coupled with the molecular detection of the rnutation(s) causative of the resultant deficiency of the GALT enzyme activity. The measurement of galactose-1-phosphate is usually monitored as well to correlate the clinical manifestations with the compliance of the patient towards galactose restricted diet. The actual methodology (Beutler method) for classic galactosemia biochemical identification proceeds via the monitoring of the fluorescence of NADPH under the relevant wavelength. The NADPH is the product of phosphoglucomutase and glucose-6-phosphate dehydrogenase catalysed reactions as illustrated in figure 2.1. All these assays are performed from a Guthrie card sample.
UDP-glucose
+
galactose-1 -phosphate UDP-galactose+
Glucose-l- phosphate(measured)
igure 2.1: The schematic representation of the Beutler method of GALT activity assay. (I), (2) and (3: i I
represent GALT, phophoglucornutase and glucosed-phosphate dehydrogenase enzymes, respectively. Adapted from G ~ n e and Straton, 1968.
2.2.2 Treatment and management of classic galactosemia
Once a conclusive diagnosis has been made, therapy in the form of dietary withdrawal of galactose is usually instituted immediately to avoid further complications. The dietary withdrawal includes breast milk and other milk based foodstuff since milk is the primary source of galactose which is a hydrolysis product of lactose. Powder milk brands such as
som mil@,
~rogestemil@ and ~ u t r a m i g e n ~ are known for their reduced galactose content and are often used as effective breast milk replacements (Chung, 1997). Dietary restriction has been shown to reverse acute symptoms such as hepatomegaly, cataracts and jaundice related to GALT deficiency (Fujimoto et a[., 2000). However, multiple organ injury occurs rapidly in classic galactosemia patients if treatment is delayed and may become irreversible if treatment is hrther delayed.It is also recorded in various literatures that despite early dietary intervention and adequate treatment, long-term outcome of galactosemia patients is less than satisfactory (Guerrero et al., 2000). There is usually primary ovarian failure in the majority of female sufferers irrespective of dietary intervention as well as neurological difficulties (Segal and Berry, 1995).
2.3. Congenital Hypothyroidism
Primary congenital hypothyroidism (CH) is usually caused by a defect in the thyroid gland itself while central hypothyroidism, although rare, is a result of a pituitary or hypothalamic defect. Congenital hypothyroidism is one of the most common causes of preventable mental retardation and affects approximately 1 newborn infant in 3000 in Canada (Simoneau-Roy et al., 2004); 1 in 3500 - 4500 in the UK and USA (Van Vliet, 2001; Fisher, 2003; Beardsall and Ogilvy-Stuart., 2004; Eugster et al., 2004). This high prevalence makes congenital hypothyroidism the most common congenital endocrine disorder. The origin of the disorder may in the minority of cases be autosomal recessive while in the majority of cases a result of a number of factors grouped together as thyroid dysgenesis. These include an ectopic thyroid with no thyroid tissue in the normal cervical position, leading to premature arrest of downward migration median thyroid during embryogenesis. Sometimes it is a result of complete absence of thyroid follicular cells or even hypoplasia of an orthotopic bilobed thyroid. Congenital hypothyroidism is almost never diagnosed clinically during the fust few weeks of life, when irreversible brain damage is occurring (Van Vliet and Czernichow, 2004).
Before the advent of biochemical screening in the early seventies, 40% of children with congenital hypothyroidism required special education mainly because of delayed diagnosis (Van Vliet, 2001; Beardsall and Ogilvy-Stuart, 2004). Congenital hypothyroidism before the mass screening era in developed countries used to be one of the more frequent causes of mental retardation in children. Introduction of screening programmes allowed early identification and treatment of congenital hypothyroidism patients, thus significantly reducing mental damage (Macchia et al., 1999).
2.3.1 Biochemical diagnosis of congenital hypothyroidism
Thyroid stimulating hormone (TSH) levels have been shown to be the most sensitive and specific screening test and these are used in programmes in the UK and Europe. Although quantification of TSH may miss the central causes of hypothyroidism, it is considered sufficiently rare not to warrant inclusion of thyroxine (T4) determination in the screening programmes of many developed countries (Beardsall and Ogilvy-Stuart, 2004). Consequently, the relative reliability of TSH measurement is often used as the preferred biochemical marker of congenital hypothyroidism rather than thyroxine (Macchia et al., 1999). Its elevation in a newborn blood sample is the earliest available laboratory manifestation of primary congenital hypothyroidism (Lott et al., 2004).
2.3.2 Treatment of congenital hypothyroidism
The aim of treatment in hypothyroidism is the normalisation of thyroid hormone levels. Thyroxine (T4) is essential for neurological development and usually by the time clinical symptoms and signs of CH have developed, irreversible damage has already occurred (Beardsall and Ogilvy-Stuart, 2004). The severity of CH does determine the outcome of intellectual impairment even though the onset of treatment has a bearing on the outcome and prognosis of the disorder. Children with severe CH who were treated earlier with a high initial dose of L-thyroxine (10 - 15 pgikglday) have been reported to manifest
normal global developmental outcome at school entry age (- 5 years) (Van Vliet, 2001; Simoneau-Roy et al., 2004). Oral sodium L-thyroxine is the treatment of choice, and there seem to be no additional benefits to the patient with triiodothyronine (T3) supplementation. Monitoring of the efficiency of the treatment is often achieved by monitoring the levels of T4 and maintain at upper range of normal T4 concentration as well as monitoring and suppressing the TSH levels to between 0.5 - 5 mU/L at regular
intervals, concurrently. However, there are risks associated with over-treatment. Therefore, early diagnosis and prompt treatment of congenital hypothyroidism is essential for the better optirnisation of neurological outcome. The commencement of therapy within the fust 13 days after birth seem to have better resolution of psychomotor development if maintained in the fust year irrespective of the severity of the disorder before therapy (Bongers-Schokking et al., 2000).
2.4. Perspective over newborn screening in the South African socio-economic environment
Population studies evaluating and monitoring the extent and severity of genetic disorders are critical if timely prevention andlor management of such disorders is a priority. With the current situation in South Africa, taking into account the ever-increasing burden of disease coupled with the difficulties surrounding the accessibility of rural communities to tertiary health facilities; alternative ways of making newbom screening for genetic disorders possible are imperative. Due to limited resources in South African state hospitals, newborn infants usually leave the hospital facilities within 24 hours of delivery unless there are presenting complications. Although 24 to 48 hour samples would be adequate for screening of hypothyroidism (Lon et al., 2004), the early departure limits the time that one can collect specimens before the mothers and their infants are discharged from the hospital. The departure of newbom babies and their mothers from the state hospitals usually happen within twenty four hours and in some cases even as little as five hours after delivery. This has a negative impact especially in identifying hypothyroidism cases since the TSH concentration levels are known to rise during and a few hours after birth process to levels reminiscent of hypothyroidism and reaches the normal childhood levels at a few months of age. Additionally, clinical manifestations of hypothyroidism during the first few weeks of life are either completely absent or non- specific (Van Vliet and Czernichow, 2004).
It is therefore imperative that results that are obtained through the measurement of TSH are interpreted with reference to expected values according to the age of the neonate.
However, TSH assays are sensitive enough to distinguish normal levels from those babies with primary congenital hypothyroidism with a cut off value of less than 15mUl1 with immunofluorometric methods and less than 20mUIL with immunoradiological assays (Beardsall and Ogilvy-Stuart, 2004). Resampling is often required after a minimum of 48 hours for infants with exceptionally high levels of TSH to confirm the absence or presence of hypothyroidism. TSH levels are known to stabilise at lower levels within 24
-
48 hours after birth if the baby's thyroid is functioning optimally (Lon et al., 2004). Although hypothyroidism has a late clinical onset, it is crucial that it is diagnosed early in life since there is no known therapy if diagnosed late. Clinical symptoms for hypothyroidism are ambiguous to diagnose even by specialist paediatricians in the early life of a neonate, newborn screening remains one of the best methods to accurately and timely diagnose hypothyroidism and avoid late onset complications.Newborn screening is better facilitated when there is enough staff in the participating health institutions to coordinate and refer the affected patients to the relevant laboratories that help in the diagnosis of the genetic disorder. However, there is a critical shortage of staff in the South African public health institutions such as clinics and even provincial hospitals. The staff to patient ratio at the Nkangala district which is one of the best staffed district in the Mpurnalanga province is 3 professional nurses to 10 000, and ranges between 0.9
-
0.003 per 1000 for the other critical health professionals such as medical officers and pharmacists (District Health System, 2004). The shortage of personnel in the public health institutions demands new alternatives to make a newborn screening programme a reality.Alternatives, as was employed in this study, include mobilising affected communities by including retired nurses to help with the screening process, by including and activating non-government organisations (NGO) to help with actual collection of required specimens, and rallying around families of the affected individuals. The collation of data regarding the birth defects in South African hospitals is well underway in major centres of South Africa. This is coordinated by the Birth Defects Surveillance System (BDSS) at Cape Town University. However, this surveillance includes only birth or genetic defects that are clinically detectable at birth as recommended by International Clearinghouse for Birth defects Monitoring System (ICBDMS). This form of data collation primarily excludes genetic disorders that may not be clinically detectable at birth (DOH, 1999). However, there are still major challenges even for a pilot newborn screening programme such as the one described in this report. These include the discharge of new mothers from the South African hospitals within 6 to 24 hours after giving birth, which then makes the chances of missing a newborn presenting with galactosemia andor hypothyroidism which manifest clinically as a late onset disorders to be increased. Any form of congenital genetic disorders monitoring and evaluation system aimed at effective prevention and management should include the collation of clinical data from all the wards wherein children are admitted according to the South African public hospital system. These include the maternity ward where babies are delivered, the paediatric ward where sick babies return to, as well as the chddren's intensive andor high care unit where critically ill babies are admitted to and cared for.
Newborn screening should be incorporated as part of the Primary Health Care (PHC) as advocated by the National Department of Health although the implementation still needs to be refined (DOH, 2001). This may include firstly, newborn screening for preventable genetic disorders in easily accessible areas. Secondly, fashioning a protocol that would define clinical indications for which specific cases may be referred for further investigation of a possible genetic disorder to laboratories with the appropriate capacity. The latter would be especially valuable for primary health care clinics in rural areas where there is usually no specialist paediatrician in attendance. The limitation regarding the inability to implement newborn screening in all health centres seem to be primarily due to economic rather than logistic constraints. Utilisation of the services of retired nurses, as was the case in this study, could also help alleviate the high workload experienced by the maternity ward nursing sisters and also facilitates the coordination and referral of children that may have clinical indications for a specific genetic disorder in paediatric wards.
Finally, with the limited funds and logistic problems in terms of accessing the majority of the rural populations, the newborn screening programme for preventable disorders in South Africa would have to use protocols that are adapted to deal with the inherent problems of a developing country.
The studies described in section 2.5. and 2.6. were entirely conceived and performed by the candidate of this thesis. The co-authors helped with the formulation of ideas and scientific editing of the manuscripts. The manuscripts will be submitted to the South African medical journal and Journal of clinical Chemistry, respectively.
2.5. Logistical challenges and cost efficiency of newborn screening for congenital hypothyroidism and classic galactosemia in a rural South African setting.
P.J. Lebea*' (MSc) and P.J. pretoriusb (DSc)
'Biosciences, Council for Scientific and Industrial Research, Ardeer Road, Modderfontein, 1645, South Africa.
b School for Biochemistry, North-West University, Potchefstroom Campus, Potchefstroom. South Africa.
Abstract
Objectives: To develop a cost efficient protocol for newborn screening of both classic galactosemia and congenital hypothyroidism and address logistical challenges for newborn screening in a rural South African setting.
Study design: Blood spot specimens were collected from over thirty percent of newborn infants out of a total of 3374 in the Nkangala district of Mpumalanga Province over a six month period. The specimens were first assayed for TSH levels as an indication of congenital hypothyroidism. The specimens were also screened for classic galactosemia using total galactose quantification and GALT enzyme activity assays. Real-time PCR was used to detect S135L galactosemia mutation which predominates in African populations.
Results: The incidence of congenital hypothyroidism was found to be 0.1% while none of the newborns presented with classic galactosemia. The S135L mutation genotyping indicated the heterozygous genotype frequency to be at 0.0079, while the mutant allelic frequency was p = 0.0039. The high recall rate in congenital hypothyroidism samples increased laboratory costs of screening by 7.3%. However, the overall cost of screening is likely to be reduced due to better accessibility to free counselling service and better sampling coordination.
Conclusion: There seems to be a higher incidence of hypothyroidism compared to classic galactosemia in the Nkangala region of South Africa consistent with global observations. Cost effective newborn screening for congenital hypothyroidism and classic galactosemia in rural South African populations can be achieved albeit requiring different communication and sample collection strategies, as well as logistical planning.
Introduction
Newborn screening is a process of testing newborn babies for treatable genetic, endocrinologic, metabolic and haematologic diseases (www.uchica~okidshosoital.org).
Screening is usually done within days of birth so that appropriate treatment can begin as soon as possible. Failure to institute therapy in the first few days of life often result in irreversible clinical damage (DeLong and Adams 1987 and Beardsall and Ogilvy 2004). Phenylketonuria is globally the most commonly screened for disorder while congenital hypothyroidism and classic galactosemia are usually incorporated as additional diseases for screening (Devi and Naushad 2004).
In congenital hypothyroidism screening, the measurement of thyroid stimulating hormone (TSH) levels andor thyroxine (T4) determines the probability of defective production of the thyroid hormones and hence forms the basis of laboratory diagnosis of hypo or hyperthyroidism (Simoneau-Roy et a1 2004). Clinical consequences of congenital hypothyroidism include mental retardation, decreased growth rate and skeletal development (LePage et a1 2004). The current treatment for hypothyroidism includes hormonal therapy with intravenous or oral supplementation with levo-thyroxine (Van Vliet 2001; Beardsall and Ogilvy 2004).
Early institution of therapy has been shown to be successful in amelioration of mental retardation as reviewed in (Gmters ef a1 2002). Classic galactosemia (OMIM 230400) is a clinically heterogeneous, autosomally recessive metabolic abnormality, which involves a galactose-1-phosphate uridyltransferase (GALT) enzyme deficiency that interrupts the normal course of galactose metabolism (Fujimoto et a1 2000). Clinical complications due to classic galactosemia include feeding problems, failure to thrive, hepatocellular damage and sepsis if untreated (Ng et a1 1994). Laboratory diagnosis for galactosemia proceeds through the demonstration of elevated erythrocyte total galactose (galactose and galactose-1-phosphate) concentration (Eu et al 1999) and consequent determination of GALT enzyme activity is used exclusively for the confirmation of classic galactosemia (Henderson et a1 2002). This measurement of total galactose can be used to screen for the deficiency of any of the three Leloir pathway enzymes (Suzuki et a1 2001; Lebea and Pretorius 2005). The current treatment includes early dietary galactose restriction to the affected individual.
Mandatory newborn screening programmes exist in many developed countries. With the exception of New Zealand and the Philipines, most developing countries do not have any screening programmes for congenital hypothyroidism or classic galactosemia (www.oaediatrics.or~.nz; www.human~enetics.co.oh). The challenges related to implementation of such programmes in developing countries often centre on the lack of financial resources and logistical problems to reach the bulk of the population, which is usually based in rural areas. The prevalence of classic galactosemia andlor congenital hypothyroidism in South Africa is still unknown.
Studies involving the evaluation of incidence of classic galactosemia on South African populations have been few and far between and the incidence of this disorder has been estimated to be between 1114400 and 1121904 (Henderson et a1 2002; Manga et a1 1999).
This study is the first of its kind to evaluate ways to save screening costs of simultaneous newborn screening protocol of congenital hypothyroidism and classic galactosemia in a mostly rural South A6ican population. Additionally, this study gives an indication of the incidence of clinical hypothyroidism and classic galactosemia in the studied population which is largely consistent with that published in the scientific literature.
Materials and Methods
Sample collection and storage
Ethical approval for this study was obtained from the North-West University (ref: 04M04) and the Mpumalanga Provincial Department of Health. Informed consent forms were signed by the mothers of the babies before samples were collected. A total of 1012
babies were sampled in a six month period. Samples were collected in the maternity, paediabic and high care wards of the Middelburg hospital by venipuncture of either side of the heel of the newborn babies. The ethanol pre-sterilised foot heel was punctured and blood was allowed to flow freely and soak onto the Guthrie card. The samples were collected at least six hours after the first breast feeding session. This sampling is usually within 24 hours after birth. The blood spot was subsequently dried for 2 to 4 hours on a
dry
clean surface at room temperature. Thereafter the cards were stored at 4 ' ~ in a sealed container and transported to the laboratory for analysis within 7 days of collection.All cards were labelled with patient information indicating the identity, relevant clinical information and more importantly the contact information of the baby's mother.
Total galactose assay
A 2.2 mm diameter dried blood discs from Guthrie filter papers were punched out and placed in 275 pL capacity conical bottomed 96-well plates. The total galactosemia kit was obtained from Bio-Rad (Rosebank, South Africa). The kit consisted of eluting, enzyme and colour reagents (cat no.: 5326053). The total galactose assay was performed on the Coda EIA analyser (Bio-Rad) using the total galactose kit according to the manufacturer's recommendations.
GALT enzyme activity assay
All specimens that exhibited higher values, that is, above the cut off value of 0.9 mg/dL for the total galactose assay, were subjected to the GALT enzyme activity assay. All samples were from Guthrie cards, and plastic bags were used to control humidity and samples were kept in the fridge (4 OC) to avoid adversely high temperatures before analysis. The GALT enzyme activity assay was performed according to the Beutler method as described in the second edition of the manual of biochemical methods by Gmne and Stratton (Beutler 1971). In this assay, the conversion of NADP' to NADPH was measured using a Perkin Elmer fluorometer set at excitation and emission wavelengths of 347nm and 4 6 h , respectively. The amount of NADH produced per unit time represented the rate of catalysis of galactose-I-phosphate breakdown by the GALT enzyme.
DNA extraction
Genomic DNA extracts suitable for PCR were prepared 6om 1.2 mm diameter punched disks as follows; the disks were treated with methanol (200
pL)
for 30 minutes and air dried for 30 minutes. Subsequently, the dried blood spots were boiled in 50 pL of distilled water. Aliquots of 10 pL were used for each PCR amplification reaction, in a total volume of 25 pL. A hot start was required to obtain consistent amplification. In samples where the amplification protocol yielded poor results an alternative method was followed. In the alternative method, the DNA was isolated from 1.2 mm diameter dried blood specimens using the FTA DNA extraction kit (cat no.: WB120061) according to manufacturer's directions (Merck). The FTA treated blood card was then used as the DNA template in the final PCR mixture without further DNA extraction in a 25 pL total volume.Common mutation genotyping
Specimens known to contain the S135L mutation were used as positive controls to validate the PCR cycle and hybridisation probes were designed as described in reference (Dobrowolski et a1 2003). In short, The PCR protocol comprised of one cycle of initial denaturation at 9 4 ' ~ for 2 minutes followed by 40 cycles with denaturation at 9 5 ' ~ for 20 seconds; annealing at 6 3 ' ~ for 20 seconds and extension at 68% for 20 seconds. The melt curve analysis was performed in two steps by first heating to 9 5 ' ~ for 10 seconds followed by cooling to 40%. Secondly, by upward temperature ramping for 110 cycles at 0 . 5 ' ~ for 15 seconds each cycle until the highest temperature of 9 5 ' ~ was reached. Fluorescence was acquired continuously during the upward ramp and computed.
The melting curve was computationally generated by plotting the 4 F I d T of the melting curve against temperature. The difference in fluorescence was continuously computed to determine the melting point of the different alleles. The quencher probes were labelled with Cy3.5 and the detector probes with fluorescein. Genotyping of the two common GALT mutations (S135L and Q188R) were performed on the Bio-Rad icyclerm. PCR was performed using the Eppendorff Real Mastermix probe Rox (cat no.: 0032002.458).
The TSH assay
The Coda EIA analyser was used together with the radioimmunoassay kit for TSH measurements (cat no.: 5312301) and the manufacturer's recommended method for TSH assay was applied without any changes (Bio-Rad). The CDC validated standards as well as the in-house standards for hypothyroidism were used for calibration as positive and negative controls for TSH levels.
Results and discussion
Sample collection
Pamphlets written in the Mpumalanga Province's native languages explaining the value of newborn screening were given to the mothers of the newborns. The information was also verbally communicated during a briefing session to all new mothers aiding in better understanding to those that cannot read. The effects of congenital hypothyroidism and classic galactosemia were explained verbally by the resident retired nurse (volunteer) to facilitate informed consent by the mothers.
Only after signing the informed consent form were the samples collected and the procedure of a possible follow-up sample explained. In our experience, the individualised session of sample collection with the retired nurse from the same community as the mothers facilitates the question and answer environment about the disorders and in turn
boosts the willingness by the mothers to submit repeat samples in time when requested. In cases of repeat samples required, the resident retired nurse in the community of choice can reach the patient by using mobile phones where applicable or by relaying the call to local taxi operators to reach remote villages.
This form of communication was surprisingly efficient since 14/81 repeat samples got the message through the local taxi call and the remainder had access to a mobile phone personally or through a relative. The mothers typically brought in the babies within 48 hours of the telephonic or taxi relayed request for re-sampling.
Total galactose assay
Only four of the 1012 (Appendix A) samples showed levels of total galactose reminiscent of galactosemia. This represents a sample collection of just over 28% of the infant population in Mpumalanga's Nkangala region. For a population of over 161 000 and investigating a disorder with an incidence of 1 in 14 000, a much bigger population would have been ideal. The reasons pertaining to the low sampling ratio to the population are related to the diverse areas accessible for maternity services. Of the four samples that were positive in terms of the total galactose assay, the repeat samples proved to be negative. It is not uncommon in this kind of an assay to get false positives especiaaly if the positive samples are marginally above the cut-off point of 0.9 mg1dL.
The absence of clinical characteristics in the newborn infants corroborated the finding that total galactose metabolites were within the normal limits =0.9 mgidL in the repeat samples. In the final analysis, all samples showed values within acceptable limits confming the absence of galactosemia in the respective newborns as shown in figure
GALT enzyme activity assay
Samples with total galactose concentration values above 0.9 mg1dL in the first round of analysis were subjected to a GALT enzyme activity assay as described in (17). The GALT enzyme activity of the selected individual specimens was found to be within the normal limits of 19.2 - 33.8 pmol/hr/gHb (Greber-Pletzer et al., 1996; Lee 2003). A second sample was then requested for all samples with total galactose levels higher than 0.9 mg/dL even if the GALT enzyme assay was negative. These specimens were only confinned negative after the second sample showed a negative GALT enzyme activity result.
Common mutation genotyping
One thousand and twelve samples were genotyped for the presence or absence of S135L mutation in GALT and eight were found to he heterozygotes. This number of heterozygotes is equivalent to a heterozygous allele frequency of 0.0079. Given that the S135L mutation is h o w n to be prevalent in black populations, the 0.79% (based on the Weinburg-Hardy law of genetic equilibrium) prevalence figure is not unexpected considering that 99% of the screened infants are of African origin and descent (Henderson et a1 2002).
However, this frequency for S135L is significantly lower from that expected for an autosomal recessive trait in Hardy-Weinberg equilibrium with a homozygous genotype frequency of 1/14 000
-
22 000 (Manga et a1 1999; Henderson et a1 2002). This may be explained in part by the small number of the assayed population although the prevalence of galactosemia is said to be even higher among black children of South Africa (Ojwang et a1 1999; Henderson et a1 2002). According to the sampled population the projected prevalence would be 1/63 500, but a much larger sample population would need to be assayed to confirm the consistency of the frequency throughout the population. Although one could extrapolate the incidence rate in this population, the short coming in this data could be the unreliability of small sample size to more accurately predict the classic galactosemia prevalence rates.Thyroid stimulating hormone assays
The TSH concentrations stabilise at lower levels in a 24 to 48 hours period after birth if the baby's thyroid is functioning normally (Ojwang et al., 1999). The cut-off value for the hypothyroidism TSH assay was maintained at 25.1 mU/L even though the sampling time was within 24 hours of birth. Since our sampling took place within 24 hours after birth the recall rate was higher (8%) than the norm ( ~ 3 % ) in similar methodology screening programmes (Amini et al., 2005). Eighty one samples had TSH values above the cut-off point, the two week follow-up assay revealed a TSH value less than 25.1 mUlL indicative of normal thyroid gland function. However, one newborn baby had a relatively high TSH value of 138mUlL as shown in figure 2.3.
These high levels of TSH concentrations persisted for over a month as confirmed by a four week follow-up sample that showed a TSH-value of 123.5 mUIL. The infant was subsequently referred for follow-up clinical checks by the resident paediatrician after the two week sample was persistently high and institution of the L-thyroxine therapy was also recommended. This poses a challenge since it is difficult to keep the babies longer than the 24 hour period in the South African public hospitals unless there are presenting clinical complications. However, the intensive explanation sessions between the parents and the volunteer health professionals helped in encouraging the parents to bring the babies within 48 hours in case of a repeat sample request.
Cost effectiveness and logistical challenges
Cost efficiency in the context of this study is based primarily on the reduction of costs necessary to successfully analyse a given specimen (Grosse 2005). Our protocol includes collaboration with a retired nursing professional within a community of choice that collects samples and stores the dried blood specimens accordingly before transporting them to the biochemical analysis laboratory. This is preceded by an educational session to teach the volunteers about congenital hypothyroidism and classic galactosemia. This facilitates the proper transfer of informed consent which forms the comer stone of rapid response when a repeat sample is required. Since the program volunteers are retired nursing professionals, they offer added value as they know the hospital setting and procedures. The volunteers were offered a fixed gratuity of R4O.00 per day irrespective of the total number of samples collected in a given day.
They live within the targeted community, the pre-counselling to the parents whose babies are screened is cost free and in the case of a repeat sample required the communication is relayed via the taxi drivers that frequent the area where the newborn infant and the mother stay. A single Guthrie card specimen is used for both tests further reducing the requirement to sample twice for the two disorders. A single piece of equipment is also used to screen for both disorders although two different kits are used for each respective disorder. The laboratory costs per sample were R11.70 for both congenital hypothyroidism and classic galactosemia. The high recall rate in congenital hypothyroidism due to early sampling which usually occurs within 36 hours after birth, the laboratory cost per sample increased by 7.3% to R12.63. However, the overall cost of screening should be considerably reduced due to free access to counselling by the resident nursing professional as well as free message relay to inaccessible rural areas.
Conclusion
The studies done on South African populations are few and far between to comprehensively reflect the extent or prevalence of galactosemia andor hypothyroidism in the relevant populations. Besides the lack of newborn screening programmes, the access and proper follow-up to rural communities remains a challenge. Even though this study is small to exhaustively determine prevalence values, with only 1012 babies sampled out of the possible 3373, it would seem hypothyroidism has higher incidence than galactosemia consistent with bigger screening programmes worldwide (Van Vliet and Czenichow 2004). Our programme applies several cost-savings strategies.
Firstly, the fact that galactosemia and hypothyroidism can be identified using the same piece of equipment (Coda EIA analyser, Bio-Rad) and from the same sample specimen greatly enhances the cost-saving practice desired for most screening programmes. The sensitivity and specificity of the electroimmuno assay are validated in-house in each IUII
using CDC standards as well as in-house developed methods. With each run the standards are used to correlate the outcome. Secondly, incorporating the real-time PCR protocol to identify commonly presenting mutations as a second tier protocol to metabolic screening has the potential to further reduce costs if the presenting population mutation profile is well established. So far, the S135L mutation is known to present higher preponderance in black populations. It is thus critical to include a second tier screening method for this mutation in a largely black population since the likelihood of this mutation being the causative factor for classic galactosemia is enhanced. However, more screening and accordingly adjusting the cut-off concentration of TSH for specimens sampled within 24 hours of birth could improve the recall rate and thus augment the cost savings potential of such a programme.
Finally, newborn screening for hypothyroidism alone has been known to be cost- effective, (Dietlein et a l , 2003; Lott et al 2004; Van Vliet and Czenichow 2004; Grosse 2005). Classic galactosemia and congenital hypothyroidism can lead to mental retardation but can be treated with satisfactory outcome. It is thus logical to assert that a newborn screening programme for both disorders that integrates disease identification biochemical methods, improves access to remote lying populations by involving the local communities via the retired professionals would greatly amplify its cost efficiency.
